Methods and systems are provided for an AdHoc enrollment process. A user may be able to enroll and be verified by a system for a variety of actions or authentications without being forced to turn over personally identifiable information and without having to formally enroll. The system may compare captured biometric information with existing biometric information and may identify the user without the use of personally identifiable information.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
2. The method of claim 1, wherein the confirmation is sent over a network to a user device.
A system and method for secure communication involves transmitting a confirmation message over a network to a user device. The confirmation message is generated in response to a verification process, which may include validating user credentials, device authentication, or transaction approval. The system ensures secure transmission by encrypting the confirmation message before sending it over the network, using protocols such as TLS or other encryption standards. The user device, which may be a smartphone, tablet, or computer, receives the confirmation and displays it to the user, providing assurance that the verification process was successful. This method is particularly useful in financial transactions, account access, or multi-factor authentication systems where secure and reliable communication is critical. The system may also include additional security measures, such as time-limited confirmations or one-time passwords, to prevent unauthorized access. The network transmission ensures that the confirmation is delivered in real-time, allowing users to proceed with their intended actions without delay. This approach enhances security while maintaining usability, making it suitable for applications requiring high levels of trust and verification.
3. The method of claim 1, wherein the first biometric data and the second biometric data comprise information captured from one or more of a facial scanner, a camera, a microphone, a fingerprint scanner, and an iris scanner.
This invention relates to biometric authentication systems, specifically improving the accuracy and reliability of user verification by analyzing multiple biometric data sources. The problem addressed is the limitations of single-source biometric authentication, which can be vulnerable to spoofing, environmental interference, or user variability. The solution involves capturing and processing biometric data from multiple input devices to enhance security and reduce false positives or negatives. The method involves collecting first biometric data from a primary input device, such as a facial scanner or camera, and second biometric data from a secondary input device, such as a microphone, fingerprint scanner, or iris scanner. These data streams are analyzed to verify the user's identity, with the secondary data serving as a cross-check to confirm the authenticity of the primary data. For example, facial recognition may be supplemented by voice recognition or fingerprint matching to ensure the user is physically present and not relying on a static image or recording. The system may also dynamically adjust the weighting of each biometric source based on environmental conditions or historical accuracy metrics to optimize performance. This approach improves security by making it harder for attackers to bypass authentication using a single compromised biometric sample. It also increases reliability by compensating for limitations in individual sensors, such as poor lighting affecting facial recognition or background noise interfering with voice analysis. The method is applicable in high-security environments like banking, healthcare, or government systems where robust identity verification is critical.
4. The method of claim 1, wherein the confirmation triggers an alarm system.
A system and method for monitoring and confirming the presence of a user in a designated area, particularly in environments where unauthorized access or absence detection is critical, such as security-sensitive facilities or automated systems. The method involves detecting a user's presence using sensors, such as motion detectors, biometric scanners, or proximity sensors, and verifying their identity through authentication mechanisms like passwords, PINs, or biometric verification. Once confirmed, the system triggers an alarm system to signal the user's presence or absence, ensuring immediate awareness of security status. The alarm system may include audible alerts, visual indicators, or automated notifications sent to monitoring personnel or connected devices. This method enhances security by providing real-time confirmation and alerts, reducing the risk of unauthorized access or undetected breaches. The system can be integrated into existing security infrastructure or deployed as a standalone solution, offering flexibility for various applications. The method ensures reliable monitoring and rapid response to security events, improving overall safety and operational efficiency.
5. The method of claim 1, wherein the confirmation indicates the user is on a whitelist.
A system and method for user authentication and access control involves verifying user identity and determining authorization status. The method includes receiving a user input, such as a biometric scan or credential entry, and comparing it against stored authentication data to confirm the user's identity. Once identity is verified, the system checks whether the user is on a predefined whitelist of authorized individuals. If the user is confirmed to be on the whitelist, access is granted to a restricted system, resource, or service. The whitelist may be dynamically updated based on administrative policies or external data sources. The method ensures secure and efficient access control by combining identity verification with authorization checks, reducing unauthorized access risks while maintaining usability. The system may be applied in physical security, digital authentication, or multi-factor authentication scenarios.
6. The method of claim 1, wherein the comparing the first template with the second template comprises using an algorithm to classify the first template.
The invention relates to template comparison techniques, particularly for classifying a first template against a second template. The method addresses the challenge of accurately determining similarities or differences between structured data sets, such as patterns, images, or other digital representations, to improve decision-making in applications like quality control, authentication, or pattern recognition. The process involves comparing a first template with a second template by applying an algorithm to classify the first template. The classification algorithm evaluates features, structures, or characteristics of the first template and matches them against the second template to determine a classification result. This may include identifying similarities, discrepancies, or hierarchical relationships between the templates. The algorithm may use machine learning, statistical analysis, or rule-based logic to perform the classification, ensuring robust and repeatable results. The method enhances accuracy in template-based systems by leveraging automated classification, reducing manual intervention and improving efficiency. This approach is useful in fields requiring precise template matching, such as document verification, biometric identification, or industrial inspection, where consistent and reliable classification is critical. The algorithm may be optimized for speed, scalability, or adaptability to different template types, ensuring versatility across applications.
7. The method of claim 6, wherein the algorithm is a Support Vector Machine (SVM).
This invention relates to machine learning systems for classification tasks, specifically addressing the challenge of improving accuracy and efficiency in predictive modeling. The method involves using a Support Vector Machine (SVM) algorithm to classify data points by identifying optimal hyperplanes that maximize the margin between different classes. The SVM is trained on labeled data to learn decision boundaries, enabling it to classify new, unseen data with high precision. The method may include preprocessing steps such as feature extraction or normalization to enhance the SVM's performance. Additionally, the SVM may be optimized using techniques like kernel functions or regularization to handle non-linear decision boundaries and prevent overfitting. The system is designed for applications where robust classification is critical, such as medical diagnosis, fraud detection, or image recognition. The use of SVM ensures scalability and adaptability to various data types, making it suitable for both small-scale and large-scale classification problems. The method may also incorporate cross-validation to evaluate model performance and fine-tune hyperparameters for optimal results. By leveraging SVM's strengths in handling high-dimensional data and minimizing classification errors, this approach provides a reliable solution for automated decision-making in diverse industries.
8. The method of claim 7, wherein the algorithm is a K-Nearest Neighbor (KNN) algorithm.
This invention relates to a machine learning method for classifying data points using a K-Nearest Neighbor (KNN) algorithm. The method addresses the challenge of accurately classifying data points in a dataset by leveraging proximity-based classification. The KNN algorithm identifies the K nearest neighbors of a target data point in a feature space and assigns the target point to the most common class among those neighbors. The method includes preprocessing the dataset to normalize or scale features, selecting an appropriate value for K, and computing distances between the target point and all other points in the dataset. The distance metric used can be Euclidean, Manhattan, or another suitable measure. The algorithm then ranks the neighbors by distance and determines the majority class among the top K neighbors to classify the target point. This approach is particularly useful in applications where labeled training data is available but the underlying data distribution is complex or non-linear. The method may also include techniques to optimize performance, such as dimensionality reduction or feature selection, to improve classification accuracy and computational efficiency.
9. The method of claim 1, wherein the threshold is set by a user device.
A system and method for dynamically adjusting a threshold value in a computing environment involves setting the threshold based on user input from a user device. The threshold is used to control a process or operation within the system, such as filtering data, triggering an action, or determining a system state. The user device, which may be a smartphone, tablet, or other computing device, allows the user to specify the threshold value directly or through an interface. The system receives the threshold value from the user device and applies it to the process, ensuring that the threshold can be customized according to user preferences or environmental conditions. This approach enables flexible and adaptive control over system behavior, improving usability and performance. The method may include additional steps such as validating the threshold value, storing it for future use, or adjusting other system parameters based on the threshold. The user device may also provide feedback or notifications related to the threshold setting, ensuring the user can monitor and adjust it as needed. This dynamic threshold adjustment enhances system responsiveness and user control in various applications, including security systems, environmental monitoring, and industrial automation.
11. The system of claim 10, wherein the first biometric modality and the second biometric modality each comprise information related to one or more of a facial scan, an iris scan, a fingerprint scan, and a voice scan.
A biometric authentication system uses multiple biometric modalities to enhance security and accuracy in user verification. The system captures and processes biometric data from at least two distinct modalities, such as facial scans, iris scans, fingerprint scans, or voice scans. Each modality provides unique biometric information that is analyzed to authenticate a user. The system may combine data from these modalities to improve recognition accuracy or use them independently for different verification purposes. By leveraging multiple biometric inputs, the system reduces the risk of false positives or negatives, making it more reliable than single-modality systems. The system is designed for applications requiring high-security authentication, such as access control, financial transactions, or identity verification. The use of diverse biometric data ensures robustness against spoofing and other fraudulent attempts, enhancing overall system security.
12. The system of claim 10, wherein the alert is sent over a communication network to a user device.
A system for alerting users about specific events or conditions involves generating an alert based on detected data and transmitting the alert to a user device over a communication network. The system includes a monitoring component that collects data from one or more sources, such as sensors, databases, or external systems. This data is analyzed to determine whether predefined conditions or thresholds are met. When a condition is detected, an alert is generated, which may include details about the event, its severity, and recommended actions. The alert is then sent to a user device, such as a smartphone, tablet, or computer, via a communication network, which may include wireless networks, the internet, or other data transmission protocols. The user device receives the alert and may display it to the user, trigger a notification, or take automated actions based on the alert content. The system ensures timely and reliable communication of critical information to users, enabling prompt responses to detected events or conditions. The communication network may support various protocols and encryption methods to ensure secure and efficient transmission of alerts.
13. The system of claim 12, wherein the alert triggers an alarm system.
A system for monitoring and alerting in a security or surveillance environment addresses the need for automated detection and response to unauthorized or suspicious activities. The system includes sensors or cameras that capture data from a monitored area, a processing unit that analyzes the data to detect predefined events or anomalies, and an alert mechanism that generates notifications when such events are detected. The alert mechanism may include visual, auditory, or digital signals to notify security personnel or other authorized users. In this specific configuration, the alert triggers an alarm system, which may include sirens, strobe lights, or other deterrent measures to respond to the detected event. The system may also integrate with existing security infrastructure, such as access control systems or emergency response protocols, to enhance situational awareness and response efficiency. The processing unit may employ machine learning or pattern recognition to improve detection accuracy over time. The system is designed to minimize false positives while ensuring timely alerts for genuine threats, making it suitable for residential, commercial, or industrial security applications.
14. The system of claim 10, wherein the first template is determined using the Neyman-Pearson lemma.
The invention relates to a system for optimizing decision-making in statistical hypothesis testing, particularly in scenarios where minimizing false positives or false negatives is critical. The system addresses the challenge of selecting optimal decision thresholds in noisy or uncertain environments, where traditional methods may fail to balance Type I and Type II errors effectively. The core of the system involves a template-based approach to hypothesis testing, where predefined templates are used to evaluate observed data against statistical models. These templates encode decision rules that determine whether to accept or reject a hypothesis based on the likelihood of the observed data under different models. A key aspect of the system is the determination of the first template using the Neyman-Pearson lemma, a fundamental result in statistical decision theory. The Neyman-Pearson lemma provides a method for constructing the most powerful test for a given significance level, ensuring that the probability of a false positive (Type I error) is controlled while maximizing the probability of correctly rejecting a false null hypothesis. By applying this lemma, the system optimizes the decision threshold to achieve the desired trade-off between false positives and false negatives. The system may also include additional templates that are determined using other statistical methods, allowing for flexibility in different testing scenarios. The overall goal is to improve the reliability and efficiency of hypothesis testing in applications such as signal detection, medical diagnostics, and quality control.
15. The system of claim 10, wherein the degree of similarity is a percent-based value.
A system for evaluating similarity between data sets or objects computes a degree of similarity expressed as a percent-based value. The system includes a comparison module that processes input data to identify matching or corresponding features between the data sets or objects. A similarity calculation module then quantifies the degree of similarity by comparing the identified features and generating a percentage that represents the proportion of matching features relative to the total features analyzed. The system may further include a normalization module to standardize the similarity measurement across different data types or scales, ensuring consistent and comparable results. The percent-based similarity value allows users to easily interpret the degree of correspondence between the data sets or objects, facilitating decision-making in applications such as pattern recognition, data matching, or quality control. The system may also include a user interface for displaying the similarity results and adjusting parameters to refine the comparison process.
17. The system of claim 16, wherein each of the session tags omits associating personally identifiable information (PII) with either of the first biometric modality and the second biometric modality.
Biometric authentication systems often require multiple biometric modalities (e.g., fingerprint and facial recognition) to enhance security. However, integrating these modalities while protecting user privacy remains a challenge, as personally identifiable information (PII) may be inadvertently linked to biometric data. This system addresses the issue by processing biometric data from two distinct modalities (e.g., fingerprint and facial recognition) without associating PII with either modality. The system generates session tags for authentication sessions, ensuring that biometric data remains decoupled from identifiable user information. These tags facilitate secure authentication without exposing PII, maintaining privacy while enabling multi-modal biometric verification. The system may also include a biometric data processor that extracts features from the modalities and a privacy controller that enforces PII omission rules. This approach ensures that biometric authentication remains robust while adhering to privacy regulations.
18. The system of claim 17, wherein a record of each of the session tags is stored in a database.
A system for managing session tags in a database is disclosed. The system addresses the challenge of tracking and organizing session-related data in a centralized repository. Session tags are metadata labels applied to user sessions to categorize or identify specific interactions, events, or attributes within those sessions. The system includes a database configured to store a record of each session tag, allowing for efficient retrieval, analysis, and management of session data. The database may be structured to associate each session tag with relevant session details, such as timestamps, user identifiers, or session attributes. This enables applications to query the database to filter, sort, or aggregate session data based on the tags. The system may also include a tagging module that dynamically generates or assigns session tags based on predefined rules or real-time session analysis. The stored records can be used for auditing, reporting, or improving user experience by analyzing patterns in tagged sessions. The database may support indexing or search functionalities to optimize performance when querying session tags. This system enhances the ability to monitor and derive insights from user sessions by maintaining a structured and searchable repository of session metadata.
20. The system of claim 19, wherein the user device is configured to at least one of decrease and increase the predetermined threshold.
A system for adjusting a predetermined threshold in a user device is disclosed. The system addresses the problem of static threshold settings in user devices, which may not adapt to varying user preferences or environmental conditions. The system includes a user device with a processor and a memory storing instructions executable by the processor to adjust the predetermined threshold. The threshold may be dynamically modified to either decrease or increase based on user input or system conditions. This adjustment allows for more flexible and responsive device operation, improving usability and performance. The system may also include a communication interface for transmitting data related to the threshold adjustment to a remote server or another device. The threshold may be associated with various device functions, such as sensitivity settings, alert levels, or operational limits, ensuring adaptability to different scenarios. By enabling dynamic threshold modification, the system enhances user control and device efficiency.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
February 26, 2021
December 20, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.